For example, some server systems (information processing apparatuses) that are used to perform business operations include a main system used to perform business operations and an independent sub-system used to perform operations different from the business operations. In the above-described server systems, a main processor provided in the main system performs processing relating to ordinary business operations on an operating system (OS) installed in the main system which will often be referred to as a main OS. On the other hand, a sub-processor provided in the sub-system performs processing relating to the initial diagnostics, the fault monitoring, etc. for hardware on an OS installed in the sub-system (hereinafter often referred to as a sub-OS).
In the above-described server system, the main processor generates OS dump data (kernel dump data) based on the contents of a memory area used by the kernel or the processor information when a fault or a panic occurs in the main system. The generated OS dump data is stored in a disk device connected to the main system. The fault analysis is performed by referring to the OS dump data stored in the disk device as a storage device.
The technology of collecting and recording memory dump data obtained when the system fault occurs on a disk shared among many nodes in a cluster system including the many nodes has been perceived as a method of collecting the OS dump data.
[Patent Document 1] Japanese Laid-open Patent Publication No. 2003-30011